关键词: 模糊C-均值聚类, 混沌粒子群, 图像分割, 逻辑自映射, 早熟收敛

Abstract: Fuzzy C-means clustering algorithm is sensitive to initial clustering center and membership matrix and likely converges into the local minimum, so it can not get the best clustering results. A new image segmentation algorithm based on chaos particle swarm and FCM clustering is proposed. The uniform particles are produced by logical self-map function.When it gets into the premature convergence,the algorithm can start the chaos optimization to improve the performance of convergence into the local minimum because of standstill. The experimental results show that the new algorithm has faster convergence and higher accuracy of segmentation.

Key words: Fuzzy C-means clustering, Chaos particle swarm,  Image segmentation,  Logical self-map, Premature convergence